Qualitative Research

Obtain a description of the quality of relationships, actions, situations, or other phenomena.

Observation, interviews and document analysis are strategies

Quantitative Design

Obtain numerical data on variables.

Emphasize prediction, generalizability, and causality.

Can be non-eperimental (collect data on variables) or experiemtnal (conduct to test hypthesis about effects)

Independent Variable

Treatment or intervention.

Must have at least 2 levels.

Alters status of the dependent variable

Dependent Variable

 Outcome which is observed and measured.

Is effected by the independent variable.

True experimental research

Provides enough control to conclude that observed variability in DV is caused by variability in IV, through control of conditions, levels of variables and RANDOM ASSIGNMENT of subjects to groups.

Quasi-experimental research

Cannot control assignment of subjects to groups and must use intact pre-existing groups or single treatment group.

Stratified Random Sampling

If population varies in specific "strata," use stratified random sampling to ensure that each stratum is represented in the sample. 

Divide according to specific strata and then randomly select subjects from each stratum (age, race, education, etc.)

Three factors that cause variability in Dependent Variable.

1. The Independent Variable (experimental variance)

2. Systematic error (error due to extraneous variables)

3. Random error (error due to random fluctuations in subjects, conditions, methods of measurement)

To be more certain that observed variability in DV is due to IV than error you should choose a design that...

1. Maximizes variability in DV due to IV

2. Control variability due to extraneous variables

3. Minimize variability caused by random error.

Techniques to control effects of extraneous variables

1. Random assignment of subjects to treatment groups.

2. Holding the extraneous variable constant.

3. Matching subjects on the extraneous variable.

4.Building the extraneous variable into the study (Blocking)

5. Statistical control of the extraneous variable

Internal validity is obtained when...

...it allows an investigator to determine if there is a causal relationship between the IV and DV.

External validity is obtained when...

...its findings can be generalized to other people, settings and conditions.

Four factors that threaten external validity

1. Interaction between testing and treatment: pretest can sensitize to purpose of the study.

2. Interaction between selection and treatment: sample should be representative to population.

3. Reactivity: people respond in a certain way because they know they are being observed.

4. Multiple treatment interference: order or carryover effects.

 Between-Groups Design

Different levels of the IV to different groups to compare the DV. 

Main effect - effect of 1 IV on the DV

Interaction - effects of 2 or more IVs considered together.

Within-Subjects Design

All levels of the IV are administered sequentially to all subjects. 

Comparisons of different levels of IV are made within subjects rather than between subjects.

Single subject design vs. group designs

1. Each single subject design includes at least 1 baseline (no treatment) phase and 1 treatment phase.  Each subject acts as his own no treatment control.

2. The DV is measured repeatedly at regular intervals throughout the baseline and treatment phases.

Reversal Designs

Includes more than one baseline.

Expansion includes the withdrawal of the treatment during the second and subsequent baseline phases

If status on DV returns to the initial baseline level during the 2nd A phase, and then to its previous treatment level during the 2nd B phase, you can be more certain change in the DV is due to IV rather than history.

Nominal Scale

Divides variables into unordered categories

If number is assigned it only acts as a label - there is no order.

Limitation: only math operation performed is to count frequency of cases in each category

Ordinal Scale

Divides observations into categories and provides info on order.

Can say one has more or leass or is midway.

Ranks or Likert scales are examples.

Limitation: cannot say 10 is twice as much as 5.

Interval Scale

Order and equal intervals between successive points on the measurement scale.

Can do addition and subtraction

No absolute zero.

Ratio Scale

Most complex.

 Has order and equal intervals and absolute zero. 

Can multiple or divide so you can say one is 3 times more than another.

Skewed Distribution

 More than half of the observations fall on one side of the distribution and a relatively few observations fall in the tail on the other side of the distribution.

Positive or negative skew.

Positive skew: most scores in negative low score side and positive tail is extended because of a few high scores.

Central Limit Theorem

As the sample size increases, the sampling distribution of the mean approaches a normal distribution.

The mean of the sample is equal to the population mean.

THe SD of the sample is equal to the SD of the population divided by the sample size.

Standard Error of the Mean

The SD of the sampling distribution.

Provides an estimate of the extent to which the mean of any one sample randomly drawn can be expected to vary from the population mean as a result of the sampling error. 

Null Hypothesis

The IV does NOT have an effect on the DV.

Alternative hypothesis

Opposite of the null hypothesis.  The IV DOES have an effect on the DV

2 Possible alternative hypotheses

1. A nondirectional (2 tailed) hypothesis means that the null hypothesis is false, but cannot determine whether it is greater or less than the parameter in the null hypothesis

2. A directional (1 tailed) hypothesis predicts whether the population parameter will be greater than or less than the null hypothesis parameter.

Alpha

The size of the rejection region is defined by alpha, which is the level of significance. 

If alpha is .05, 5% of the sampling distribution is the rejection region and the remaining 95% s the retention region.